There are two ways for a gap to grow. The top can pull away, or the bottom can fall out. Here in the United States, the key problem is that the bottom is falling, and the changes are not small. At the national level, in fourth grade reading, the scores of the top 10% of students fell 0.5 points from 2013-24. Meanwhile, the scores of the bottom 10% fell 15 points. 

The state-level results are even more jarring. To put those in perspective, consider that the average student gains about 8 points per year on the National Assessment of Educational Progress reading tests. The chart below shows the change from 2013 to 2024 in the scores for the bottom 10% of students in each state. For this group, 40 states saw a decline of 10 points or more, 16 saw declines of 20 points or more and two states — Delaware and Maryland — had declines of more than 30 (!) points.

But one state is bucking this trend: Mississippi. Indeed, there’s been a fair amount of coverage of Mississippi’s reading progress in recent years, but its gains are so impressive that they merit another look.  

First, it’s worth remembering that Mississippi is the state in the country. Its per-capita income is below $50,000, and it spends less on its public schools than all but three states. But when the Urban Institute NAEP scores based on each state’s demographics, Mississippi’s fourth-grade reading scores came out on top. 

Even without those adjustments, though, Mississippi looks pretty great. Its Black students third nationally, and its low-income kids outperform those in every other state. Mississippi is also the only state to see gains across all performance levels over the last decade. Its average went up, but so did the scores of its highest and lowest performers. Mississippi raised the bar and the floor at the same time. 

So how did they do it? How did Mississippi go from 49^th in the country a decade ago to near the top today? And what can other states learn from it? 

According to a recent piece by Grace Brazeale, a policy associate with the advocacy group Mississippi First, the state implemented of changes starting with the 2013 . That law funded the state department of education to hire, train and literacy coaches to the 50 lowest-performing schools. It also required schools to administer universal screenings to identify students with reading deficiencies early and to communicate those results to parents, and it required schools to hold back students who were not reaching a certain threshold by third grade. 

These changes were not all that expensive, but they had big effects. EdWeek’s Elizabeth Huebeck in 2023 that the state spent $15 million per year to support its literacy work, and 60% of that went to coaching and intervention staff. A research paper last fall from Noah Spencer from the University of Toronto that the law helped drive the state’s gains.

Spencer estimated that the third-grade retention policy alone could be responsible for about one-quarter of the gains, and it was surely the most controversial element. Some have even tried to cast doubt on Mississippi’s NAEP gains by arguing they’re merely a function of testing older kids. But this has been : Mississippi does hold back more kids than other states, but it , and the average age of Mississippi’s NAEP test-takers has barely budged over time. 

on third-grade retention policies has that students who are retained tend to have better outcomes than those who are not, but that the process for identifying those children can be biased against Black, Hispanic, and low-income students. 

However, I think what matters most is not the students who are retained, but what the policy does to . Mississippi required schools to notify parents when their child was off track and to craft individual reading plans for those with reading deficiencies. In other words, the threat of retention may have shifted behavior in important ways. As evidence for that theory, consider that a study out of Florida also found positive effects on younger siblings of students who are retained. 

Other states have literacy policies too, but they are often weaker than what Mississippi implemented. For example, a 2023 paper from Michigan State University researchers John Westall and Amy Cummings that 41 states and the District of Columbia had early literacy policies, but only 12 could be considered . Many states had the superficial elements of a literacy plan, but they lacked requirements that all districts adopt high-quality instructional materials, screen all students for dyslexia and take dramatic action for kids who continue to struggle to read. Critically, Westall and Cummings found that only states with truly comprehensive policies saw student learning gains. 

This is perhaps the key lesson for state policymakers: The only literacy policies that are likely to lead to significant student learning gains are ones that meaningfully change schools and classrooms. With reading scores continuing to fall in many parts of the country, policymakers should look to replicate the lessons from Mississippi. 

The examined data from 2011-12 to 2016-17 and found that students who were retained in third grade scored about 18 points higher in English language arts and math in fourth grade than low-performing peers who were not retained. The gains continued through seventh grade, though at a slower rate. 

“I was surprised at the huge … positive effect,” said NaYoung Hwang, an assistant professor of education policy at the University of New Hampshire, who co-authored the report with Cory Koedel, a professor of economics and public policy at the University of Missouri.

The study is unusual because it compares third graders who were retained with peers who scored just high enough to advance. These groups are expected to be statistically similar, Hwang added. 

Despite this progress, the number of Indiana students held back after third grade has been declining for years, sliding from 2.4%, or 1,535 students, in 2013 to 1.25%, or 762, in 2016, according to Hwang. 

Those 762 students were among 3,500 who did not pass the Indiana Reading Evaluation and Determinations, also known as IREAD3, in 2016 after being allowed to retake the exam over the summer. This year, nearly 5,000 students failed the test. State officials did not answer a request for information about how many of them repeated third grade.

The policy “hasn’t been uniformly applied anymore,” said Bob Behning, a state representative who chairs the education committee.

The study found no increases in absenteeism or discipline problems among students who were held back and showed positive benefits regardless of the student’s race, gender or family socio-economic status, Hwang added. 

She acknowledged that research focused on sixth- or eighth-graders who were retained has found lower graduation and higher dropout rates. “We need more studies and years” to determine if third-graders who are held back would follow this pattern, she added. 

The Indiana Reading Evaluation and Determinations, also known as IREAD3, requires third graders to score 446 or higher to pass. Students who don’t must be offered remediation and a chance to retake the test during the summer. State law also allows for what are called good-cause exemptions, which allow some third graders who don’t hit the required score to advance anyway. These include English learners, special education students and those who have already been retained twice.

Hwang said she undertook the study because retention is a much-debated policy. More than half the states allow or require schools to retain low-performing students after third grade, but the number actually held back seems to be shrinking. In Florida, for example, the retention rate went from 15% in 2002 to just 6% in 2010, according to a by the RAND Corporation.

In Tennessee, where a new law threatened to hold back thousands of third graders, who scored low enough were actually held back. That’s 1.2%. Nearly 4,400 students were granted waivers to advance to the fourth grade after their parents appealed.

Reading experts said the research on retention is mixed. “We’re not seeing long-term benefits,” said Danielle Dennis, dean of the University of Rhode Island’s College of Education. “There’s been very little research [tracking students] into high school.”

She wondered if gains in Indiana came, in part, because such a small percentage of students were held back. Paying for an additional year of school and offering remediation would be harder with a significantly larger group, she said. 

“This [new report] gives us new info, not all the info,” Dennis said. 

“Overall, the weight of research evidence continues to be that there are not long-term benefits of retaining students,” said Nell Duke, executive director of the Center for Early Literacy Success and a professor of education and psychology at the University of Michigan. 

However, she said research shows the benefits of summer programs and small-group reading interventions, noting that these measures can be used even when students are not retained. 

Even the students’ gains in seventh grade may not be enough to justify retention, she added, as funding an additional year of school, pulling children away from their peers and possibly increasing their chances of dropping out may not be worth an extra 7 points in reading. 

“I would characterize retention as not a high-priority policy move to improve long-term literacy. There are ways to spend funds and energy that are likely to have more lasting results,” Duke added. 

She also disputed the notion that the end of third grade marks a key change for students from learning to read to reading to learn: “Today’s standards expect children to learn from reading well before third grade, and students can continue to develop as readers, with proper instruction, well after fourth grade.”

Behning noted that Indiana has shifted its priorities about reading instruction in the past two years. A law this year created an approved science of reading curriculum list, offered literacy support for elementary schools where fewer than 70% of students pass IREAD3 and provides grants to teachers and schools that improve students’ reading skills. 

In 2022, the state and the Lilly Endowment combined to put together a in literacy education, the state’s largest expenditure in that area. The funds, in part, helped pay for instructional coaches in about 200 elementary schools, as well as targeted support for students who need the most help improving reading skills, and offered a stipend of up to $1,200 to K-3 teachers who participate in professional development in the science of reading. 

, circulated as a working paper through the National Bureau of Economic Research, focuses on a Florida policy that has to boost achievement among young learners. In so doing, it adds a new wrinkle to an evidence base that has not only expanded substantially over the last few years, but also : grade retention, at least for low-scoring children in early grades, meaningfully improves their test scores. 

How that improvement is accomplished is still up for debate. While some believe schools foster the learning gains through extra instruction, others point to the simple advantages of children studying the same material after undergoing a year of cognitive and social development. And the latest results — referred to as “spillover effects” on younger brothers and sisters — raise further questions as to how retention works.

Umut Özek, a senior economist at the RAND Corporation and co-author of the latest Florida paper, said that while the academic growth he measured was likely attributable to multiple causes, families and educators could be motivated by the “threat effect” of students being flagged to repeat a grade.

“When you have this goal set in third grade, such that you need to score above a certain level to be promoted, it provides a clear signal to schools and parents that they need to do something in earlier grades so their students aren’t retained,” he remarked.

However promising the research outcomes, grade retention remains one of the most contentious planks of the education reform agenda. Since 2013, over two dozen states either allowing or requiring school districts to make grade promotion decisions based on elementary reading performance. But parents have with the policy, with some of third-grade reading exams. Legislators in and significantly relaxed their elementary reading mandates earlier this year. In Tennessee, which adopted its own retention policy in 2021, 60 percent of third graders this spring, and in an effort to demonstrate proficiency.

In the wake of generational learning loss stemming from COVID-related school closures, some experts to gradually affect larger numbers of K–12 students. Katharine Strunk, the dean of the University of Pennsylvania’s Graduate School of Education, said that the prospect of seeing their children fall short of promotion was having an “eye-opening effect” on many families.

“One thing that’s come out of the pandemic is that we know that parents are not always made aware of the challenges facing their kids at school,” Strunk said. “Maybe for the first time, parents are being told, ‘Your kid is really struggling in a way that’s much worse than his or her peers.’”

A ‘very clear signal’ for parents

Florida’s reading retention law, first enacted in 2002 under then-Gov. Jeb Bush, to score above the minimum achievement level on a literacy exam in order to move onto the third grade — though “good-cause” exemptions are often granted to children who receive special education or English learner services, who have already been held back, or who can demonstrate reading proficiency via other means. Due in part to energetic lobbying from Bush and his think tank, ExcelinEd, a slew of other states over the last decade.

To reveal the impact of the original policy, Ozek and his collaborators gathered a comprehensive set of student-level data from 12 anonymous school districts, including standardized test scores, special education status, demographic indicators and teacher characteristics. That information was combined with birth records from the same areas, allowing the researchers to link the progress of older students targeted for retention with that of their closest younger siblings. 

The paper encompasses the first seven years of the state’s grade retention system and subsequent test scores for both older and younger siblings through 2011–12; during that period, Florida’s portion of third graders retained was approximately 10 percent, though the annual rate declined from 15 percent in 2002 to just 6 percent in 2010.

Comparing kids who placed below the retention cutoff score against those who placed above it, the team found that repeating the third grade was associated with a statistically significant increase in state test scores in both reading and math. That finding largely echoes into retention in Florida, including co-authored by Ozek. 

To account for the growth, Ozek cited the breadth of resources that schools are required to provide children who are not promoted. Such students are assigned to highly effective teachers, receive 90 minutes of dedicated reading instruction each day and are given the option of attending an intensive, literacy-oriented summer camp.

“These students receive substantial support in the following year, and that support is more personalized and tailored toward their needs,” Ozek said. “That’s probably a key element behind the success of some of these policies.”

That observation echoes the conclusions of by a pair of researchers at Michigan State University. Their analysis, which examined the effects of early literacy policies on both state test scores and performance on the National Assessment of Educational Progress, showed that achievement growth was greatest in — including some form of retention, but also extensive assistance for affected students and coaching for their teachers.

“Altogether, these results indicate that the full set of interventions available under early literacy policies is important in improving literacy achievement and skills,” the authors wrote. (Strunk, until recently an education professor at Michigan State, helped review the paper.)

Beyond the direct improvements, however, Ozek’s Florida paper finds that retained students’ younger siblings also saw a bump in test scores compared with the brothers and sisters of children who were promoted to the fourth grade. That advancement was measured at approximately 30 percent the size of the main effects; it was also particularly concentrated among boys, as well as immigrant families and those including a disabled child. 

The authors offer several theories to explain why younger siblings experienced positive, albeit smaller, movement. Among them: Having an older sibling held back was correlated with being assigned to a classroom with relatively higher-performing peers, perhaps because parents of retained third-graders influenced the classroom placements of their younger children. 

Additionally, in instances where retained students attended schools that received a state accountability score lower than an A over the preceding two years, their parents were more likely to move younger siblings to schools with better reading results, higher-performing teachers (as measured by value-added scores on state tests) and those specializing in reading instruction. 

Ozek said he could understand why having a child repeat a grade would seize parents’ attention. A father of two, he noted that retention was a much starker message than performance on state tests, and one that would likely cause adults to take notice.

“It’s really hard for me, even as an education policy researcher, to assess what those [state test] scores mean,” he said. “But when you get a signal that says, ‘Your kid is not performing at a level that will allow them to be promoted to fourth grade,’ that’s a very clear signal that will likely induce a response from parents, and schools as well.”

The harms of being held back

Little outside evidence exists to either validate or undermine the new study’s claims around spillover effects on younger family members. But competing explanations have recently cast doubt on the fairness and effectiveness of retention in early grades. 

Mississippi’s third-grade “reading gate,” which largely resembles Florida’s law, has been nationally lauded for pushing up historically dismal literacy scores over the last decade. At the same time, argue that is due to a form of statistical sleight of hand; since a healthy portion of the state’s fourth-graders have been held back, their additional year of intellectual maturity — not the effects of retention and supplemental instruction — could be responsible for their progress. (Advocates have responded that while that claim may be applicable elsewhere, it is , where the “reading gate” appears not to have increased the average age of fourth graders.)

Meanwhile, studies of students retained in higher grades have found that being held back in middle or high school makes students and significantly more likely to be . While younger children are seemingly less fazed by repeating a year, the practice may be salient, and damaging, with transitions to middle and high school.

The University of Pennsylvania’s Strunk, who has of Michigan’s now-weakened retention system, agreed that students who repeated third grade had “more opportunity to develop and learn,” theoretically allowing them to achieve at higher levels on that basis alone. Beyond that possibility, she added, there is something of a paradox in sending low-achieving students back to the same classrooms and teachers that failed them the first time around.

“Is it really a good idea to give kids an extra year of school if the schooling is not working the way we want it to?” Strunk wondered. “It’s like the Einstein quote: ‘The definition of crazy is doing the same thing over and over again and expecting a different result.’”

Leaving aside the precise causes of post-retention growth, however, the bulk of the recent research suggests that retaining floundering readers can produce notable short- and medium-term gains. Another of a Florida-style literacy standard, focused on schools in Indiana, showed that third graders who scored just below the threshold for promotion ended up significantly out-performing their classmates who were narrowly promoted. Those effects extended into the middle school grades, with no sign that retention increased disciplinary or attendance problems.   

Cory Koedel, a co-author of that study and an economics professor at the University of Missouri, said he was agnostic about which explanation for the progress mattered most, or even whether the effects would eventually fade out.

“In my view, whether it’s the extra year of instruction or the extra year of maturity that’s allowing them to catch up isn’t that important. What’s important is that they’re catching up.”